Workloads that run on Solid State Drives (SSDs) are diverse, but many have real-time requirements. On-Line Transaction Processing (OLTP) is an example workload which is highly latency sensitive. For such applications, it is imperative that the latency of responses from the storage subsystem be as low as possible and time-bounded, so as not to exceed a few milliseconds. However, erase operations due to garbage collection (GC) in SSDs get in the way of achieving this requirement. This problem is referred to as the “tail latency problem,” where the end user is noticeably impacted even if a very small percentage (much less than 1%) of the reads suffers unduly large latencies.
A need remains for a way to minimize the impact of garbage collection operations on other operations of an SSD.